cusideabelincoln
Diamond Member
- Aug 3, 2008
- 3,268
- 11
- 81
One thing these sites don't test is how frametimes are affected with slower CPUs.
I just upgraded from a 2500k to a 3770k, particularly to improve Overwatch performance. And the difference in smoothness is amazing. Here are my settings and performance results. Now if you ask me why do I run such low settings? Well I tried epic settings and I was able to maintain above 60 fps, but there was terrible mouse lag. So I have to maintain a framerate above 100 fps for the mouse lag to be unnoticeable. I also use 120 Hz with lightboost/strobing, so ideally I want to maintain above 120 fps, otherwise there is sometimes a ghosting effect.
The 2500k was at 4.6 Ghz while in these tests the 3770k is at stock. I'm using an R9 280x @ 1080p
As you can see, in both tests I maintain the same average framerate. However the frame time variation is much better with the 3770k, and my bottom framerate doesn't drop as low.
You want to know how I tested? I made a custom match on Nepal. Populated the server with Bots. Played the match for a few minutes until my bot teammates and I pushed the enemy back to their spawn. Respawn timer was decreased and ultimate charge rate increased. I basically kept killing them as they walked out of spawn and used Pharah's ultimate 3 times per benchmark run. This test is as repeatable as you can possibly make it, plus it's also demanding when you have bots popping their ultimates at the same time.
I just upgraded from a 2500k to a 3770k, particularly to improve Overwatch performance. And the difference in smoothness is amazing. Here are my settings and performance results. Now if you ask me why do I run such low settings? Well I tried epic settings and I was able to maintain above 60 fps, but there was terrible mouse lag. So I have to maintain a framerate above 100 fps for the mouse lag to be unnoticeable. I also use 120 Hz with lightboost/strobing, so ideally I want to maintain above 120 fps, otherwise there is sometimes a ghosting effect.
The 2500k was at 4.6 Ghz while in these tests the 3770k is at stock. I'm using an R9 280x @ 1080p
As you can see, in both tests I maintain the same average framerate. However the frame time variation is much better with the 3770k, and my bottom framerate doesn't drop as low.
You want to know how I tested? I made a custom match on Nepal. Populated the server with Bots. Played the match for a few minutes until my bot teammates and I pushed the enemy back to their spawn. Respawn timer was decreased and ultimate charge rate increased. I basically kept killing them as they walked out of spawn and used Pharah's ultimate 3 times per benchmark run. This test is as repeatable as you can possibly make it, plus it's also demanding when you have bots popping their ultimates at the same time.
Last edited: